Uploaded image for project: 'MariaDB Server'
  1. MariaDB Server
  2. MDEV-6074

Decouple BGL code from OQGRAPH engine code and improve related documentation

Details

    Description

      The OQGRAPH code backup is still at a working prototype level of maturity, in terms of documentation and tidiness. This makes it difficult to debug and extend.

      I intend to gradually cleanup the file structure and pull out the boost graph code to a subdirectory and document how it works, so that it will become easier to implement new algorithms.

      Attachments

        Activity

          Ok, currently OQGraph is MariaDB_PLUGIN_MATURITY_BETA in 10.0. If you'd like to decrease it to MariaDB_PLUGIN_MATURITY_ALPHA or MariaDB_PLUGIN_MATURITY_EXPERIMENTAL. Or — later — increase to MariaDB_PLUGIN_MATURITY_GAMMA or MariaDB_PLUGIN_MATURITY_STABLE please, feel free to do so.

          serg Sergei Golubchik added a comment - Ok, currently OQGraph is MariaDB_PLUGIN_MATURITY_BETA in 10.0. If you'd like to decrease it to MariaDB_PLUGIN_MATURITY_ALPHA or MariaDB_PLUGIN_MATURITY_EXPERIMENTAL . Or — later — increase to MariaDB_PLUGIN_MATURITY_GAMMA or MariaDB_PLUGIN_MATURITY_STABLE please, feel free to do so.
          andymc73 Andrew McDonnell added a comment - - edited

          (I wasnt even aware of those categories - where are they documented?)

          BETA is probably fine - the interface is pretty much bedded down. Having a much improved regression test suite means this activity can be done at much lower risk than previously.

          I just want to rearrange (specifically, split up files into more understandable chunks) and document with Doxygen or something. Probably over a period of months when I have spare hours here and there.

          This should not affect the interface with MariaDB (other than to potentially mitigate latent bugs)

          andymc73 Andrew McDonnell added a comment - - edited (I wasnt even aware of those categories - where are they documented?) BETA is probably fine - the interface is pretty much bedded down. Having a much improved regression test suite means this activity can be done at much lower risk than previously. I just want to rearrange (specifically, split up files into more understandable chunks) and document with Doxygen or something. Probably over a period of months when I have spare hours here and there. This should not affect the interface with MariaDB (other than to potentially mitigate latent bugs)

          (e.g. I know Arjen is keen to see a Depth First Search implemented at some point)

          andymc73 Andrew McDonnell added a comment - (e.g. I know Arjen is keen to see a Depth First Search implemented at some point)

          People

            andymc73 Andrew McDonnell
            andymc73 Andrew McDonnell
            Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved:

              Git Integration

                Error rendering 'com.xiplink.jira.git.jira_git_plugin:git-issue-webpanel'. Please contact your Jira administrators.